Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Dummy/Effects coding, alpha draws

I used HB to estimate utilities for my CBC. I am a little confused about the difference between effects coding and dummy coding and how this results in differences in alpha draws csv output.

In my study, I have two conditions, a Control condition and a Treatment condition. I set condition as a covariate so the resulting csv file has Intercepts and Condition(1) for each part-worth utility. I would like to do Bayesian Tests specified here: https://legacy.sawtoothsoftware.com/169-support/technical-papers/cbc-related-papers/2033-statistical-testing but again, am confused about what the values of the alpha draws are since I have two Conditions but only partworths for 1?

Another thing I have noticed is that different technical papers say to ignore draws prior to convergence but then range from saying ignore the first 5K, 10K or 20K. Is there a more specific approach to determining convergence so that I can choose which number of draws to ignore?
asked Sep 2, 2020 by anonymous

Your solution to the original question

Please only use this to answer the original question. Otherwise please use comments.
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
Anti-spam verification:

To avoid this verification in future, please log in or register.